1 / 31

CS 188: Artificial Intelligence Fall 2007

CS 188: Artificial Intelligence Fall 2007. Lecture 19: Decision Diagrams 11/01/2007. Dan Klein – UC Berkeley. Recap: Inference Example. Find P(W|F=bad) Restrict all factors No hidden vars to eliminate (this time!) Just join and normalize. Weather. Forecast. U. Decision Networks.

amelia
Download Presentation

CS 188: Artificial Intelligence Fall 2007

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. CS 188: Artificial IntelligenceFall 2007 Lecture 19: Decision Diagrams 11/01/2007 Dan Klein – UC Berkeley

  2. Recap: Inference Example • Find P(W|F=bad) • Restrict all factors • No hidden vars to eliminate (this time!) • Just join and normalize Weather Forecast

  3. U Decision Networks • MEU: choose the action which maximizes the expected utility given the evidence • Can directly operationalize this with decision diagrams • Bayes nets with nodes for utility and actions • Lets us calculate the expected utility for each action • New node types: • Chance nodes (just like BNs) • Actions (rectangles, must be parents, act as observed evidence) • Utilities (depend on action and chance nodes) Umbrella Weather Forecast

  4. U Decision Networks • Action selection: • Instantiate all evidence • Calculate posterior joint over parents of utility node • Set action node each possible way • Calculate expected utility for each action • Choose maximizing action Umbrella Weather Forecast

  5. U Example: Decision Networks Umbrella = leave Umbrella Weather Umbrella = take Optimal decision = leave

  6. U Example: Decision Networks Umbrella = leave Umbrella Weather Umbrella = take Forecast =bad Optimal decision = take

  7. U Value of Information • Idea: compute value of acquiring each possible piece of evidence • Can be done directly from decision network • Example: buying oil drilling rights • Two blocks A and B, exactly one has oil, worth k • Prior probabilities 0.5 each, mutually exclusive • Current price of each block is k/2 • Probe gives accurate survey of A. Fair price? • Solution: compute value of information = expected value of best action given the information minus expected value of best action without information • Survey may say “oil in A” or “no oil in A,” prob 0.5 each = [0.5 * value of “buy A” given “oil in A”] + [0.5 * value of “buy B” given “no oil in A”] – 0 = [0.5 * k/2] + [0.5 * k/2] - 0 = k/2 DrillLoc OilLoc

  8. Value of Information • Current evidence E=e, utility depends on S=s • Potential new evidence E’: suppose we knew E’ = e’ • BUT E’ is a random variable whose value is currently unknown, so: • Must compute expected gain over all possible values • (VPI = value of perfect information)

  9. U VPI Example MEU with no evidence Umbrella MEU if forecast is bad Weather MEU if forecast is good Forecast Forecast distribution

  10. VPI Properties • Nonnegative in expectation • Nonadditive ---consider, e.g., obtaining Ej twice • Order-independent

  11. VPI Scenarios • Imagine actions 1 and 2, for which U1 > U2 • How much will information about Ej be worth? Little – we’re sure action 1 is better. A lot – either could be much better Little – info likely to change our action but not our utility

  12. Reasoning over Time • Often, we want to reason about a sequence of observations • Speech recognition • Robot localization • User attention • Medical monitoring • Need to introduce time into our models • Basic approach: hidden Markov models (HMMs) • More general: dynamic Bayes’ nets

  13. Markov Models • A Markov model is a chain-structured BN • Each node is identically distributed (stationarity) • Value of X at a given time is called the state • As a BN: • Parameters: called transition probabilities or dynamics, specify how the state evolves over time (also, initial probs) X1 X2 X3 X4

  14. Conditional Independence • Basic conditional independence: • Past and future independent of the present • Each time step only depends on the previous • This is called the (first order) Markov property • Note that the chain is just a (growing) BN • We can always use generic BN reasoning on it (if we truncate the chain) X1 X2 X3 X4

  15. Example: Markov Chain 0.1 • Weather: • States: X = {rain, sun} • Transitions: • Initial distribution: 1.0 sun • What’s the probability distribution after one step? 0.9 rain sun This is a CPT, not a BN! 0.9 0.1

  16. Mini-Forward Algorithm • Question: probability of being in state x at time t? • Slow answer: • Enumerate all sequences of length t which end in s • Add up their probabilities …

  17. Mini-Forward Algorithm • Better way: cached incremental belief updates sun sun sun sun rain rain rain rain Forward simulation

  18. Example • From initial observation of sun • From initial observation of rain P(X1) P(X2) P(X3) P(X) P(X1) P(X2) P(X3) P(X)

  19. Stationary Distributions • If we simulate the chain long enough: • What happens? • Uncertainty accumulates • Eventually, we have no idea what the state is! • Stationary distributions: • For most chains, the distribution we end up in is independent of the initial distribution • Called the stationary distribution of the chain • Usually, can only predict a short time out

  20. Web Link Analysis • PageRank over a web graph • Each web page is a state • Initial distribution: uniform over pages • Transitions: • With prob. c, uniform jump to a random page (dotted lines) • With prob. 1-c, follow a random outlink (solid lines) • Stationary distribution • Will spend more time on highly reachable pages • E.g. many ways to get to the Acrobat Reader download page! • Somewhat robust to link spam • Google 1.0 returned the set of pages containing all your keywords in decreasing rank, now all search engines use link analysis along with many other factors

  21. Most Likely Explanation • Question: most likely sequence ending in x at t? • E.g. if sun on day 4, what’s the most likely sequence? • Intuitively: probably sun all four days • Slow answer: enumerate and score …

  22. Mini-Viterbi Algorithm • Better answer: cached incremental updates • Define: • Read best sequence off of m and a vectors sun sun sun sun rain rain rain rain

  23. Mini-Viterbi sun sun sun sun rain rain rain rain

  24. Hidden Markov Models • Markov chains not so useful for most agents • Eventually you don’t know anything anymore • Need observations to update your beliefs • Hidden Markov models (HMMs) • Underlying Markov chain over states S • You observe outputs (effects) at each time step • As a Bayes’ net: X1 X2 X3 X4 X5 E1 E2 E3 E4 E5

  25. Example • An HMM is • Initial distribution: • Transitions: • Emissions:

  26. Conditional Independence • HMMs have two important independence properties: • Markov hidden process, future depends on past via the present • Current observation independent of all else given current state • Quiz: does this mean that observations are independent given no evidence? • [No, correlated by the hidden state] X1 X2 X3 X4 X5 E1 E2 E3 E4 E5

  27. Forward Algorithm • Can ask the same questions for HMMs as Markov chains • Given current belief state, how to update with evidence? • This is called monitoring or filtering • Formally, we want:

  28. Example

  29. Viterbi Algorithm • Question: what is the most likely state sequence given the observations? • Slow answer: enumerate all possibilities • Better answer: cached incremental version

  30. Example

  31. Real HMM Examples • Speech recognition HMMs: • Observations are acoustic signals (continuous valued) • States are specific positions in specific words (so, tens of thousands) • Machine translation HMMs: • Observations are words (tens of thousands) • States are translation positions (dozens) • Robot tracking: • Observations are range readings (continuous) • States are positions on a map (continuous)

More Related